SCIENCE ADVANCES | RESEARCH ARTICLE

EVOLUTIONARY BIOLOGY Copyright © 2021 The Authors, some rights reserved; The evolution of mammalian brain size exclusive licensee J. B. Smaers1,2*, R. S. Rothman3, D. R. Hudson3, A. M. Balanoff4,5, B. Beatty6,7, D. K. N. Dechmann8,9, American Association 10 11,12,13 14 6,15,16,17 18 19 for the Advancement D. de Vries , J. C. Dunn , J. G. Fleagle , C. C. Gilbert , A. Goswami , A. N. Iwaniuk , of Science. No claim to 14,20 12 21,22,23,24 25 2,3,26 W. L. Jungers , M. Kerney , D. T. Ksepka , P. R. Manger , C. S. Mongle , original U.S. Government 1 23,27 28 29 8,9 F. J. Rohlf , N. A. Smith , C. Soligo , V. Weisbecker , K. Safi Works. Distributed under a Creative Relative brain size has long been considered a reflection of cognitive capacities and has played a fundamental Commons Attribution role in developing core theories in the life sciences. Yet, the notion that relative brain size validly represents selec- NonCommercial tion on brain size relies on the untested assumptions that brain-body allometry is restrained to a stable scaling License 4.0 (CC BY-NC). relationship across species and that any deviation from this slope is due to selection on brain size. Using the largest fossil and extant dataset yet assembled, we find that shifts in allometric slope underpin major transitions in mam- malian evolution and are often primarily characterized by marked changes in body size. Our results reveal that the largest-brained mammals achieved large relative brain sizes by highly divergent paths. These findings prompt a reevaluation of the traditional paradigm of relative brain size and open new opportunities to improve our under-

standing of the genetic and developmental mechanisms that influence brain size. Downloaded from

INTRODUCTION It has long been recognized that brain size scales with body size The brain is directly responsible for governing an animal’s interac­ following a standard linear allometric power law (6). The scaling tions with its environment. As such, the brain is often considered to coefficient (slope) of this allometry is assumed to be relatively stable flexibly respond to selection in changing environments (1–3). Brain across vertebrate classes and orders (most often estimated as be­ http://advances.sciencemag.org/ size is, however, also commonly accepted to be restrained by ener­ tween 2/3 and 3/4) (7) and is thought to reflect universal energetic getic requirements that are considered universal across all verte­ growth constraints (4, 5). Largely because of methodological limita­ brates (4, 5). This apparent paradox highlights that brain size is one tions in phylogenetic comparative statistics, this working hypo­ of the most salient traits for understanding the fundamental balance thesis has received little scrutiny. Previous studies have therefore between adaptability and constraint in evolution. Despite this im­ mostly been limited to comparing residual variation along a stable portance, crucial aspects related to the timing, pattern, and drivers slope [i.e., mean relative brain size or encephalization quotient that underlie modern phenotypic diversity in brain size remain (EQ), quantified through differences in the intercept of the evolu­ undescribed. tionary allometry] (7, 8). There is, however, evidence to suggest that changes in the slope (quantifying changes in brain-body covaria­ 1 tion) may constitute an important additional source of comparative

Department of Anthropology, Stony Brook University, Stony Brook, NY 11794, USA. on April 28, 2021 2Division of Anthropology, American Museum of Natural History, New York, variation (9–12). 3 NY 10024, USA. Interdepartmental Doctoral Program in Anthropological Sciences, Identifying the points at which evolutionary shifts in brain-body Stony Brook University, Stony Brook, NY 11794, USA. 4Department of Psychologi- cal and Brain Sciences Johns Hopkins University, Baltimore, MD 21218, USA. 5Division covariation occur is of paramount importance to understanding the of Paleontology, American Museum of Natural History, New York, NY 10024, USA. selection pressures that may be operating. Whereas shifts in inter­ 6NYIT College of Osteopathic Medicine, Old Westbury, NY 11568, USA. 7United cept address changes in mean among traits, shifts in slope address States National Museum, Smithsonian Institution, Washington, DC 20560, USA. 8 changes in variation among traits (e.g., stabilizing selection restrains Department of Migration, Max-Planck Institute of Animal Behavior, 78315 Radolfzell, Germany. 9Department of Biology, University of Konstanz, 78464 Konstanz, Germany. variation, divergent selection allows for variation) (13). Failing to 10Ecosystems and Environment Research Centre, School of Science, Engineering account for the possibility that trait covariation may differ among 11 and Environment, University of Salford, Manchester M5 4WX, UK. Division of Bio- groups of species could potentially hide crucial sources of variation logical Anthropology, University of Cambridge, Cambridge CB2 3QG, UK. 12Behav- ioral Ecology Research Group, Anglia Ruskin University, Cambridge CB1 1PT, UK. that contribute to explaining phenotypic diversity. Moreover, evo­ 13Department of Cognitive Biology, University of Vienna, Vienna 1090, Austria. lutionary allometries (allometries quantified across species) are de­ 14Department of Anatomical Sciences, Stony Brook University, Stony Brook, NY termined by ontogenetic and static allometries (across developmental 11794, USA. 15Department of Anthropology, Hunter College, New York, NY 10065, USA. 16PhD Program in Anthropology, Graduate Center of the City University of time in the same individual and individuals of the same species, New York, 365 Fifth Avenue, New York, NY 10016, USA. 17New York Consortium in Evo- respectively) and thus are indicative of the genetic and developmen­ lutionary Primatology, New York, NY 10065, USA. 18Department of Life Sciences, Nat- tal mechanisms that regulate growth (14). Consequently, more de­ 19 ural History Museum, London SW7 5BD, UK. Department of Neuroscience, tailed information on the allometric patterns that characterize the University of Lethbridge, Lethbridge, AB T1K-3M4, Canada. 20Association Vahatra, BP 3972, Antananarivo 101, Madagascar. 21Bruce Museum, Greenwich, CT 06830, USA. brain-body relationship across evolutionary time will provide new 22Department of Ornithology, American Museum of Natural History, New York, NY opportunities to investigate the nature of the processes that shape 23 10024, USA. Division of Science and Education, Field Museum of Natural History, those patterns. Last, the occurrence of shifts in slope would indicate Chicago, IL 60605, USA. 24Department of Paleobiology, Smithsonian Institution, Washington, DC 20013, USA. 25School of Anatomical Sciences, Faculty of Health that comparisons of relative brain size (and EQ) are only valid among Sciences, University of the Witwatersrand, Johannesburg, South Africa. 26Turkana groups with a similar slope. The ubiquitous approach of quantify­ Basin Institute, Stony Brook University, Stony Brook, NY 11794, USA. 27Campbell ing only residual variation along a stable slope may therefore lead to Geology Museum, Clemson University, Clemson, SC 29634, USA. 28Department of 29 biased results and incorrect interpretations. Anthropology, University College London, London WC1H 0BW, UK. College of Science and Engineering, Flinders University, Bedford Park, SA 5042, Australia. Birds and mammals are of particular interest in this context be­ *Corresponding author. Email: [email protected] cause they both independently evolved relatively larger brains than

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 1 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE other vertebrate classes (7). This innovation was likely facilitated by boundary (~23 Ma ago), after which several significant shifts oc­ an easing of the phenotypic integration between brain size and body curred rapidly (Table 1 and table S2). These shifts from the ances­ size (15). Such decoupling leads to increased variation available to tral grade (slope: b = 0.51) resulted in new slopes for colobines selection, which, in turn, is expected to heighten flexibility in re­ (b = 0.67), cercopithecines (b = 0.43), lesser apes (b = 0.32), great sponse to selection (16). Recent work has shown that shifts in slope apes (b = 0.23), hominins (b = 1.10), and callitrichines (b = 0.58). are paramount to explaining the brain’s evolutionary diversification Other shifts in slope near the Paleogene-Neogene boundary oc­ in birds (12), demonstrating that the selective response to increased curred in bears and pinnipeds (Fig. 1; Table 1 and table S2), which variation is not restricted solely to changes in mean relative brain are noteworthy for having the largest downward shifts in slope in size but may also play out in terms of changes in brain-body covari­ the sample: from b = 0.58 in other carnivorans to b = 0.39 and ation. In mammals, shifts in brain-body covariation have been sug­ b = 0.23, respectively. See the Supplementary Results for a complete gested to occur in primates (9), carnivorans (17), marsupials, (18), description of allometric repatterning across clades. and among mammalian orders (11). However, it has remained un­ clear where and when such shifts occur throughout mammalian evolution and how they contribute to explaining variation in the DISCUSSION brain-body relationship. Different evolutionary trajectories for the Several methodological innovations (19, 20), in conjunction with largest-brained mammals ever-increasing data availability, make it possible to test the widely Five mammalian groups (elephants, great apes, hominins, toothed

held assumption that the slope of brain-body allometry is stable in whales, and delphinids) attained their position at the top of the Downloaded from mammals. Here, we use bivariate Bayesian multipeak Ornstein-­ bivariate brain-body space (Fig. 2) by following an unexpectedly Uhlenbeck (OU) modeling (19, 21) in combination with phyloge­ diverse range of trajectories. Elephants represent the simplest case, netic analysis of covariance (20) to identify changes in both slope as they evolved directly from the ancestral mammalian grade and and intercept of the evolutionary allometry of the brain-body achieved large relative brain size by vastly increasing their body size relationship in mammals. We apply these methods to the largest while increasing brain size even more rapidly (table S3). In toothed taxonomic sampling to date, comprising 107 extinct species and whales and delphinids, relative brain size increased in a stepwise http://advances.sciencemag.org/ 1311 extant species spanning 21 orders (table S1; we quantify size manner. The first intercept shift occurred in the stem fereuungu­ in terms of mass for both brain and body). Our approach allows us lates, which follow a trajectory similar to (although less pronounced to identify where and when allometric shifts occur in mammalian than) that in elephants, by increasing brain size more than body evolution and whether these shifts are driven primarily by changes size. This was followed by an intercept shift in stem toothed whales, in brain or body size. Our results provide new insight into the types which decreased brain and body size relative to stem cetaceans, with of selection that have shaped extant diversity and open new oppor­ body size decreasing more rapidly than brain size (uncertainty sur­ tunities for research into the underlying mechanisms. rounding this scenario exists and is a function of the interpretation of the fossil record; see the Supplementary Results). Delphinids show a third intercept shift driven by body size decrease and brain RESULTS size increase relative to other toothed whales. The evolutionary tra­ Allometric patterning through time jectory of great apes and hominins was the most complex, starting on April 28, 2021 Across more than 1400 species, mammalian brain-body allometry with two consecutive downward shifts in slope (while increasing comprises 30 distinct grades (F21,2 = 29.02, P < 0.001; L.Ratio = 525.68, both brain and body size) in stem apes and stem great apes, fol­ P < 0.001, AIC = 488, AICѡ > 0.999; Figs. 1 and 2, Table 1, and lowed by the most marked increase in slope observed in this study, table S2). The ancestral mammalian grade has a slope of 0.51 and is in hominins (from b = 0.23 to b = 1.10). Delphinids and hominins, retained by several early radiating orders (golden moles and ten­ which converge at the apex of the brain-body space, are the only recs, elephant shrews, dugongs and manatees, hyraxes, sloths, and two grades with negative brain-body correlations, that is, brain size armadillos), as well as by tree shrews, hares and rabbits, squirrels, increased while body size decreased (table S3). All five of the mam­ flying lemurs, and tarsiers (Fig. 1). Shifts in slope are common (of mal groups with the largest brain size may have originated in the 29 grade shifts, 16 include a shift in slope) and characterize both Neogene. This remarkable diversity in evolutionary trajectories and early and late diversification. Most early slope shifts occurred near late attainment of peak relative brain size parallels patterns in birds the Cretaceous-Paleogene boundary (K-Pg; ~66 million years (Ma) in which the largest-brained taxa (parrots and corvids) attained ago; Fig. 1), and all indicate a shift toward a higher slope. This tem­ large brain sizes during the Neogene via very different trajectories poral clustering suggests that changes in the relative growth trajec­ (12). These parallel patterns between birds and mammals suggest tory of brain and body size were fundamental for mammalian that, similar to the K-Pg transition, the Paleogene-Neogene transition diversification in the wake of the K-Pg mass extinction. This aligns may have created conditions ripe for ecological radiations and niche with a pattern recently observed in birds (12), suggesting that eco­ expansions that affected brain size evolution. logical radiation and subsequent niche expansion following the K-Pg mass extinction played a major role in shaping the trajectories Similar evolutionary trajectories for the by which both birds and mammals became the largest-brained ver­ smallest-brained mammals tebrate classes. In contrast to relatively large-brained clades, species that lie near Shifts in slope (both increases and decreases) were also crucial in the bottom of the brain-body space are consistently characterized later diversifications, with one prime example being anthropoid by a shift toward a higher slope and a lower intercept. Such shifts primates (Fig. 1). Stem and early crown anthropoids retained the ancestral occur in afrosoricids (b = 0.66), hystricomorph rodents (b = 0.66), mammalian condition until shortly after the Paleogene-Neogene myomorph and castorimorph rodents (b = 0.57), bats (b = 0.66),

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 2 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE Downloaded from http://advances.sciencemag.org/ on April 28, 2021

Fig. 1. Time-calibrated phylogeny of mammals with branch colors corresponding to the 30 significantly different allometric grades identified in this study (Table 1). The ancestral mammalian grade is indicated in gray, with warmer colors (green and red) assigned to higher-slope grades, and colder colors (blue and purple) to lower-slope grades. For each grade, a lighter color hue indicates grades with a lower intercept, and a darker hue indicates grades with a higher intercept (Table 1). Ar- rows indicate changes in mean body size (white arrows) or mean brain size (black arrows) resulting in grade shifts, with double arrows indicating one of these variables is changing faster than the other after considering allometry. Triangles indicate changes in cross-species trait variance in body size (white triangles) or brain size (black tri- angles), with normal triangles indicating increase in mean variance and inverted triangles indicating decrease in mean variance (tables S3 to S5). The equality sign (=) indicates no discernible change in brain size variance. See data S2 for individual species labels. Illustration by J. Lázaro. eulipotyphlans (b = 0.62), and marsupials (b = 0.58); all of which mean brain size are common (12). This effect may be rooted in the derive directly from the ancestral mammalian grade (b = 0.51) scalability of mammalian neocortical architecture. While the mammalian (Table 1). Bats, eulipotyphlans, and afrosoricids converge on the neocortex (dorsal pallium) is organized as an outer layer of neurons same grade (i.e., same intercept and slope) as the myomorph and surrounding scalable white matter, the bird dorsal pallium is orga­ castorimorph rodents and marsupials. These allometric changes are nized in a nuclear manner that might limit its scalability (22). mostly explained by a disproportionally higher decrease in mean brain size relative to body size and a lower variance in body size relative to Divergent versus stabilizing selection in brain and/ brain size (tables S3 to S5). This suggests that body size becomes or body size more restrained than brain size at smaller body sizes, permitting A detailed account of changes in evolutionary allometry across deep smaller animals to evolve disproportionately small brain sizes. time provides opportunities for understanding the types of selec­ The higher evolutionary flexibility of mean brain size relative to tion operating in certain taxa. A crucial aspect in this regard are mean body size is apparent across all mammals, with stem toothed putative differences in the cross-species variance of a trait among whales being a notable exception (table S3). This contrasts with groups. Although variance has long been considered crucial to un­ birds, where pronounced changes in mean body size compared with derstanding the types of selection operating in certain taxa (13) and

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 3 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE Downloaded from http://advances.sciencemag.org/ Fig. 2. pGLS regressions for each of the grades. The ancestral mammalian grade is indicated in each display to provide an evolutionary context. Numbers indicate the value of the slope for each grade. Colors and silhouettes are as in Fig. 1. to play a principal role in driving shifts in allometric slope between slope in the sample (b = 0.23, which they share with pinnipeds), grades (fig. S1), it was undescribed by previous research. Several whereas hominins have the highest slope (b = 1.10). Although both patterns revealed by our analyses (table S5) demonstrate how mean brain size and mean body size are similar across these two changes in the cross-species variance of a trait between groups af­ nested grades, the variance in hominin brain size is 6.45 times greater fect shifts in slope and may reveal possible selective pressures. than in great apes, while the variance in body size is only 1.58 times Stem cercopithecoids derived from the ancestral mammalian greater. In this case, a major shift in variance occurred in brain size grade by disproportionately increasing mean brain size relative to while body size remained mostly static, suggesting divergent selection mean body size and disproportionately decreasing body size vari­ on brain size (scenario displayed in fig. S1C) and more stabilizing on April 28, 2021 ance relative to brain size variance (tables S3 to S5). This pattern selection on body size (likely associated with the shift to bipedality resulted in an increase in slope (from b = 0.51 to b = 0.67; Fig. 1 and in hominins). Table 1). Within cercopithecoids, cercopithecines (baboons, ma­ caques, and relatives) diverged toward a lower slope (b = 0.43) Allometric shifts do not always represent shifts in cognition through a moderate increase in mean brain and body size (~1.28 times Variation in relative brain size is traditionally associated with cog­ greater mean), a strong increase in body size variance (3.45 times nitive capacities and behavioral flexibility, but this notion has rested greater), and a stabilization of brain size variance (1.16 times greater). on several fundamental assumptions. First, it has been assumed that Brain size of cercopithecines thus varies across a much wider range the slope of the brain-body relationship is stable across species and of body size than other cercopithecoids (i.e., colobines), suggesting that deviations from this allometry reflect selection on brain size. As more divergent selection on body size (scenario displayed in fig. S1B). a result, numerous studies geared toward explaining the evolution Considering the effects of locomotion on body size (23), and the fact of relative brain size have focused on the evolution of brain size and that cercopithecines include both arboreal and terrestrial species while cognition (25, 26). Our results do not support this assumption. Evo­ colobines are predominantly arboreal, this allometric repatterning lutionary shifts in brain-body allometry commonly included changes is likely associated with their differences in locomotor diversity. in slope and were often driven by changes in body size. Rather than A similar scenario plays out in pinnipeds, which underwent a focusing on residual deviation from a common slope, the emphasis significant decrease in slope compared with other carnivorans (from should be on the allometric shifts themselves. In addition, address­ b = 0.58 in carnivorans to b = 0.23), primarily due to decreased ing factors that are not directly tied to brain size or cognition likely brain size variance relative to body size variance (table S5). Body plays an important role. Some selective pressures that play a key size in pinnipeds thus varies widely compared with other carniv­ role in species diversification are more directly tied to body size orans given their range of brain sizes (table S5). This suggests diver­ than brain size. gent selection on body size, most likely influenced by the transition A prime example is locomotion. Known to influence body size to a semiaquatic niche (24). and energetic expenditure [e.g., high cost of flying in bats (27)], ma­ A contrasting scenario is provided by great apes and hominins, jor transitions in locomotion allow for a redistribution of energetic which have extremely different slopes. Great apes have the lowest allocation to growth, thereby providing opportunities for allometric

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 4 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE

Table 1. Phylogenetic generalized least-squares parameters for all grades identified in the analysis. Grade numbers (1 to 6) indicate groups of clades with significantly different slopes (clades with the same grade number indicate slopes that are not significantly different from each other). Within each grade with a similar slope, grade letters indicate clades with a significantly different intercept. b and a refer to the values for the slope and intercept. “Lower” and “Upper” refer to the lower and upper bounds of the 95% confidence intervals. Note that some grades with low sample size are not listed here because they did not include a significant shift in slope (only a significant shift in intercept). These grades include elephants (n = 8), Cebus (n = 6), Atelinae (n = 8), Saki/Uakari (n = 4), Daubentonia (n = 1), Tragulina (n = 4), and pangolin (n = 1). These grades are, however, listed in table S3, which analyzes their mean brain and body sizes. NW, New World; OW, Old World. Slope Intercept Grade Clade n b SE Lower Upper a SE Lower Upper 1A Pinnipeds 32 0.226 0.080 0.064 0.389 3.001 1.045 0.872 5.130 1B Stem great apes 7 0.229 0.061 0.085 0.373 3.520 0.657 1.967 5.073 2A Stem apes 9 0.377 0.159 0.018 0.736 1.394 1.475 −1.944 4.731 2A Ursids 10 0.323 0.077 0.151 0.496 1.922 0.801 0.137 3.707 2A Cercopithecines 56 0.426 0.033 0.360 0.493 0.675 0.312 0.050 1.301 3A Ancestral 130 0.510 0.021 0.469 0.551 −1.588 0.357 −2.293 −0.882 3B Stem NW monkeys 16 0.402 0.051 0.294 0.510 0.309 0.383 −0.503 1.121 Downloaded from 3B Stem fereuungulates 100 0.542 0.020 0.501 0.582 −1.028 0.295 −1.612 −0.443 3D Stem-toothed 31 0.468 0.035 0.396 0.540 1.093 0.436 0.204 1.982 whales* 3E Delphinids 16 0.533 0.046 0.435 0.631 0.847 0.550 −0.320 2.013 http://advances.sciencemag.org/ 4A Dunnarts 15 0.545 0.092 0.349 0.741 −2.645 0.251 −3.180 −2.111 4B Castorimorphs/ myomorphs 180 0.567 0.014 0.538 0.595 −2.226 0.130 −2.482 −1.969 4B Stem marsupials 150 0.580 0.018 0.544 0.615 −2.140 0.174 −2.484 −1.797 4C Strepsirrhines 56 0.577 0.031 0.514 0.639 −1.403 0.267 −1.938 −0.869 4C Callitrichines 16 0.578 0.059 0.452 0.704 −1.279 0.361 −2.044 −0.515 4C Stem carnivorans 178 0.577 0.019 0.539 0.614 −1.666 0.204 −2.068 −1.264 4D Stem 29 0.671 0.074 0.520 0.822 −1.676 0.673 −3.052 −0.300 cercopithecoids°

5A Eulipotyphlans 48 0.618 0.029 0.560 0.676 −2.712 0.158 −3.030 −2.394 on April 28, 2021 5A Stem bats 217 0.672 0.015 0.642 0.701 −2.834 0.068 −2.968 −2.700 5A Afrosoricids 13 0.659 0.072 0.503 0.815 −3.010 0.369 −3.808 −2.213 5B Hystricomorphs 19 0.663 0.047 0.565 0.760 −2.674 0.357 −3.421 −1.926 5B OW fruit bats 47 0.665 0.016 0.633 0.697 −2.366 0.071 −2.510 −2.223 6A Hominins 11 1.097 0.164 0.736 1.458 −5.304 1.722 −9.095 −1.514

*“Stem toothed whales” refer to stem and crown nondelphinid toothed whales. °“Stem cercopithecoids” consists of the extinct species Victoriapithecus and crown colobines.

repatterning. This is also apparent in birds, where relative brain size to a semiaquatic niche in pinnipeds (AIC = 31.527; AICѡ < 0.001) is associated with flight style and complexity (28). Our analyses coincides with disproportionate decrease in brain size variance rel­ confirm the influence of locomotion on brain-body allometry in ative to body size variance. Notably, the initial transition to an aquatic mammals by demonstrating that many major transitions in loco­ niche in cetaceans is not linked to a shift in brain-body allometry, motion coincide with allometric repatterning events. For example, although later shifts to higher mean brain size do occur in toothed the evolution of flight in bats coincides with an allometric shift in whales and delphinids. This makes whales a compelling counter ex­ intercept and slope (AIC = 69.339; AICѡ < 0.001) that is charac­ ample, despite being largely freed from size constraints and reach­ terized by decreased body size variance, suggesting strong con­ ing the largest size of any vertebrates, baleen whales retain the same straints on body size due to the highly specialized and energetically brain-body allometry as their terrestrial cetartiodactylan relatives. demanding locomotion in this group. The transition from the Within primates, allometric shifts coincide with increased commit­ mammalian ancestral grade to terrestrial cursorial locomotion in ment to terrestriality in cercopithecines (AIC = 81.709; AICѡ < fereuungulates (AIC = 46.051; AICѡ < 0.001) coincides with a dis­ 0.001). The origins of the great ape grade coincide with a decreased proportionate increase in mean brain size to body size. The transition slope (AIC = 14.632; AICѡ < 0.005) and is associated with a

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 5 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE substantially larger increase in body size variance compared with In general, these results indicate that the brain-body relationship brain size variance. This may similarly have been associated with reveals more than just selection on brain size. Therefore, relative increased reliance on orthograde behaviors related to terrestriali­ brain size may not always be a valid proxy of cognition. The same ty (29). The transition to bipedalism in hominins coincides with a argument applies to the widely used EQ measure (8), which is also marked increase in slope (AIC = 58.527, AICѡ < 0.001) that is quantified using deviations from a stable slope. A possible way to characterized by higher brain size variance relative to body size improve the comparative study of cognition is to compare different variance. brain regions. Whereas comparisons among brain regions associated Other factors that likely influence the brain-body allometry pri­ with different functions would reveal neurobehavioral specializa­ marily through changes in body mass include, but are not limited tions (36, 38), comparisons among brain regions from different de­ to, sexual size dimorphism, diving depth in aquatic niches (30), velopmental precursors would highlight changes in growth allocation antipredator defensive mechanisms (31), and energetic strategies to (10). Such remapping factors ensure validity by using hypotheses maintain homeostasis (possibly contributing to explaining the allo­ that are based on established neuroanatomical and neuroscientific metric shifts in the convergent eulipotyphlans and afrosoricids) principles (40). Comparisons among brain regions also have the po­ (32). Overall, the general alignment of allometric shifts with major tential to reveal which patterns of brain region evolution explain transitions in locomotion (table S2) and concomitant selection on brain size evolution (41) and whether such patterns of brain region body size (e.g., small body size in a volant niche, large body size in evolution can be tied to cognition (38). Such analyses are essential terrestrial and aquatic niches) suggests that the evolutionary repat­ because the evolution of brain size may not always be in line with

terning of the brain-body relationship reflects an adaptive profile the evolution of brain regions (or other neuroanatomical features) Downloaded from that extends beyond selection on brain size alone (33). The over­ that are associated with higher cognition. The association between whelming focus on brain size and cognition in the literature there­ increased brain size and increased complexity is assuredly strong fore should be reconsidered. (42), but crucial exceptions to this trend suggest that much may be A second fundamental assumption is that brain-body allometry left to discover on this topic. For example, whereas some archaeo­ reflects the maintenance of basic autonomic, and sensory functions cetes (fossil stem cetaceans) had brains that are larger than toothed and allometric deviations therefore reflect cognition (34). This in whales (data S2), their brains have relatively smaller cerebral hemi­ http://advances.sciencemag.org/ turn relies on the assumption that there is little variation in the rel­ spheres compared with toothed whales (43). Early cetacean brain ative volumes of brain regions and that larger brains are mostly evolution may thus have comprised two different trajectories: in­ scaled up small brains (35). If true, the only target for explanation creased brain size with low complexity in archaeocetes, and stable would be the (allometrically adjusted) brain-to-body ratio, irrespec­ or decreased brain size with high complexity in toothed whales. In tive of how changes in body size alter allometric patterning. This this example, complexity does not match absolute brain size, al­ assumption has, however, already been disproven by numerous though it does match with relative brain size as toothed whales have multivariate studies of brain region sizes (36, 37). A poignant com­ a higher relative brain size. In the above described comparison of parison in this context is that between the polar bear (Ursus maritimus) the polar bear and the California sea lion (38), however, complexity and the California sea lion (Zalophus californianus). Although these does not match either absolute or relative brain size (or EQ). Both two species have a similar mean body size, the brain size of the polar these empirical cases confirm that deviations from the general asso­ bear is twice that of the California sea lion. Consequently, the polar ciation between brain size and complexity occur and may be a source on April 28, 2021 bear exhibits a much higher (four times higher) relative brain size. of future discovery. Such trends are of paramount importance to However, the California sea lion has 3.6 times more volume devoted the study of cognition and can only be revealed through compari­ to brain regions that are associated with higher cognition relative to sons among brain regions (or other neuroanatomical features). regions associated with basic autonomic and sensory functions (38). This is likely associated with vocal learning and other cogni­ Allometric shifts reveal comparative differences tive skills in the California sea lion (39). Although cognitive tests in in adaptive profile polar bears are generally lacking, this example indicates that relative The primary importance of identifying evolutionary allometric brain size alone may not be a sufficient proxy for the amount of shifts lies in the fact that they provide fundamental information on brain tissue that is allocated to higher cognition (33). both the patterns and the processes that shape extant variation. Be­ This example further illustrates the potentially confounding cause evolutionary allometries are determined by ontogenetic and effects of assuming a stable slope across all species. As mentioned, population-level allometries, they can be considered as macroevo­ pinnipeds display the lowest slope among mammals, a pattern that lutionary signatures of changes in the microevolutionary mechanisms is most likely driven by divergent selection on body size. Given that that regulate growth (14). Accurate identification of macroevolu­ pinnipeds are large mammals, such a low slope inevitably results in tionary shifts thus provides crucial information on the mechanisms low relative brain size when considering a common slope (the com­ that shape comparative differences in adaptive profile. mon slope is higher than that observed in pinnipeds alone). Attrib­ The potential of this approach is arguably best exemplified by uting this low relative brain size to selection on brain size and humans, where a shift to bipedality allowed for a redistribution of cognition obscures the trajectory that resulted in their low relative energy from locomotion to reproduction and brain growth (44). brain size (namely, most predominantly selection on body size, not This redistribution of energy to the brain effected changes in the brain size). Overall, assuming a stable slope across all species im­ mechanisms of its growth, for example, delayed expression of genes plies that selection on body size is comparable across species—an associated with synaptic development (45) and neotenic changes in assumption that is difficult to uphold given the wide variety of niches mRNA expression (46) for those brain regions that explain human that mammals inhabit and the fundamental role that body size plays brain size expansion (36, 41). In turn, these developmental changes in ecological and evolutionary processes. caused an evolutionary allometric shift that is characterized by a

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 6 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE significantly higher ratio of brain variance to body variance, which for the complexity and diversity of the underlying processes and ul­ indicates divergent selection on brain size (causing an increase in timately encapsulating aspects beyond cognition and brain size. slope; see Figs. 1 and 2 and table S5). Humans represent an evolutionary allometric shift that is driven primarily by changes in brain size. We observed other allometric MATERIALS AND METHODS repatterning events driven primarily by brain size in elephants, Old Data World fruit bats, toothed whales, and delphinids. In other clades Data on brain and body size (both quantified as mass) were gleaned evolutionary allometric shifts may be primarily driven by selection from the literature (table S1 and data S2). Comparisons between for body size (e.g., cercopithecines, great apes, and pinnipeds). In all techniques using brain tissue mass data and endocranial volume cases, these allometric shifts occur at transitions into a new niche data from the skulls of fossil species have been validated in multiple and/or changes in energetic requirements or energetic availability published studies (48, 49), and endocranial volume has been proven and further involve redistribution of growth allocation, shifts in the a reliable proxy estimate of brain size of both mammalian and non­ genetic and developmental mechanisms that regulate growth, and mammalian taxa (11). The phylogeny is a consensus tree derived by allometric repatterning. Smaers et al. (38) from the mammalian supertree compiled by Although the mechanisms underlying different types of allometric­ Faurby et al. (50). Fossil placement was done according to the liter­ shifts are not yet fully understood, we emphasize that accurate identifi­ ature (table S1) and is detailed in data S2. cation of these shifts is a necessary first step toward reaching this goal.

Contrary to the traditional paradigm of relative brain size, we show Identifying shifts in allometric patterning Downloaded from that allometric shifts can be characterized by changes in slope and may Methods are similar to those reported in previous work (12). We be caused by changes in both brain and/or body size. This prompts a estimated differences in slope and intercept of the brain-to-body reevaluation of the conventional concept of a grade shift as only rep­ relationship directly from the data using a Bayesian multiregime resenting changes in intercept and reveals that a full understanding of OU modeling approach (19). The OU model assumes that the evo­ the evolution of brain size relative to body size requires the consider­ lution of a continuous trait “X” along a branch over time increment ation of effects that extend beyond selection on brain size alone. “t” is quantified as dX(t) = [ − X(t)]dt + dB(t) (51). Relative to http://advances.sciencemag.org/ the standard Brownian motion (BM) model [dX(t) = dB(t)], the Implications for the use of “relative brain size” OU model adds parameters that estimate mean trait value () and Relative brain size and the related EQ measure are one of the most the rate at which changes in mean values are observed (). The in­ widely used measures in comparative biology and have played a clusion of these additional parameters allows an appropriate differ­ fundamental role in developing several core theories in the life sci­ entiation between changes in the mean ( and ) and variance () ences (4, 47). Here, we demonstrate that the way in which relative of a trait over time and thus renders the OU model framework more brain size and EQ are traditionally quantified (using deviations appropriate than BM for modeling changes in the direction of trait from a stable slope) may result in erroneous inferences on which evolution. Here, we used a bivariate implementation of OU model­ taxa increased or decreased brain size and hampers a deeper under­ ing that is explicitly geared toward estimating shifts in slope and standing of the patterns and types of selection that explain changes intercept of evolutionary allometries by using reversible-jump Mar­ in brain size (and body size). In other words, our results demon­ kov chain Monte Carlo (MCMC) machinery (21) (“OUrjMCMC”). on April 28, 2021 strate that the traditional statistical measures of relative brain size We implemented this approach by combining 10 parallel chains of and EQ do not always validly capture variation in brain size. We 2 million iterations each with a burn-in proportion of 0.3. We al­ demonstrate that a more nuanced approach to quantifying varia­ lowed only one shift per branch, and the total number of shifts was tion of brain size relative to body size (quantifying changes in both constrained by means of a conditional Poisson prior with a mean the intercept and the slope of the evolutionary allometry, combined equal to 2.5% of the total number of branches in the tree and a max­ with investigating univariate patterns of change in brain and/or imum number of shifts equal to 5%. Starting points for MCMC body size that underpin these bivariate changes in intercept and chains were set by randomly drawing a number of shifts from the slope) provides new insights and opens new opportunities for im­ prior distribution and assigning these shifts to branches randomly proving our understanding of the patterns and processes that char­ drawn from the phylogeny with a probability proportional to the acterize brain size evolution. size of the clade descended from that branch. The MCMC was ini­ In general, our results do not contradict the notion that variation tialized without any birth-death proposals for the first 10,000 gener­ in brain size is associated with cognition. Our results rather demon­ ations to improve the fit of the model. The output of this procedure strate that the traditional measures of relative brain size and EQ do generates an estimate of a best-fit allometric model with posterior not always validly capture variation in brain size. This result cau­ probabilities assigned to each shift in slope and/or intercept. tions against the unequivocal use of relative brain size or EQ to In part due to difficulties in parameter estimation intrinsic to quantify or study cognition. We argue that the evolution of cogni­ OU modeling (52), the bivariate OUrjMCMC output may include tion is more validly represented by comparisons among brain re­ false positives and/or false negatives (21). To identify false nega­ gions (or other neuroanatomical features). Such comparisons have tives, we ran a univariate OU model estimation procedure (19) on the potential to identify which patterns of brain region evolution the residuals of each grade to detect shifts in mean. If such shifts in explain brain size evolution and therefore reveal more precise and mean were detected, they were added as shifts in intercept to the relevant information regarding the evolution of cognition (38, 41). allometric model (no such shifts were detected for these data). To This does not render the study of brain size relative to body size identify false positives, the allometric model was translated to a useless. On the contrary, it reframes this trait more broadly to rep­ least-squares framework and used in a confirmatory analysis using resent comparative differences in adaptive profile, thereby accounting phylogenetic analysis of covariance (“pANCOVA”) (20). Although

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 7 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE pANCOVA uses a different evolutionary process than OU modeling expected change in mean brain size relative to change in mean body (i.e., BM instead of OU), it is expected that grade membership as esti­ size of 0.47 with a maximum expectation of 0.55 (table S3). The mated by the OU modeling is confirmed using least-squares analysis. mammalian ancestral grade has a mean brain size of 1.92 and a Because BM assumes fewer statistical parameters, pANCOVA can be mean body size of 6.92 (log scale). The stem cercopithecoid/colo­ considered as a conservative confirmatory test of the significance of bine grade has a mean brain size of 4.40 and a mean body size of grade membership as estimated by OU modeling. All reported results 9.04. The difference in mean brain size from the mammalian ances­ are those that were confirmed by pANCOVA (Table 1 and table S2). tral grade to the colobine grade is +2.48; that of body size is +2.12. _¯​ brain ​ The ratio (​​ ​ ​) ​​ is thus 1.17, which is 0.62 points above the up­ ​ ​​ ¯​ body ​​ ​​ ¯​ body​ Assessing differential changes in mean brain and/ ​​​ ​​ ​ _ ​​ ​​​​ per bound brain-to-body expectation (table S3). The ratio ( ¯​ brain)​ is or body size 0.86, which is 0.88 below the upper bound of the body-to-brain ex­ To assess whether changes in the brain-to-body allometry were pectation (table S4). Colobines thus indicate more change in mean driven primarily by mean brain size or body size, phylogenetic brain size relative to change in mean body size than expected from means for both brain size and body size were calculated for each of their ancestral grade. the allometric grades identified by the allometric patterning analy­ Stem toothed whales are the only grade in the sample that indi­ sis. Phylogenetic means were calculated following standard phylo­ cates more change in mean body size than change in mean brain genetic generalized least-squares procedures (20). size. Relative to stem cetaceans (archaeocetes), stem toothed whales Patterns of mean brain/body size increase/decrease were evalu­ decrease in size (although note the uncertainties inherent in this

ated by comparing mean differences in brain size and body size be­ inference discussed in the Supplementary Results). Archaeocetes Downloaded from tween ancestral and descendant grades (or “derived grades”; note have a mean brain size of 7.25 and a mean body size of 15.03. Stem that we, here, consider “descendant” and “derived” as equivalent toothed whales have a mean brain size of 6.67 and a mean body size terms) (table S3). The ratio of the difference in ancestral-to-descendant of 11.90. The differences in mean brain and mean body size from mean brain size to the difference in ancestral-to-descendant archaeocetes to stem toothed whales are thus −0.58 and − 3.13, re­ ¯brain​ ​ _​​ ​​ mean body size (¯​ body​; log scale) was considered as an indication of spectively. The confidence interval of the slope for the ancestral grade _¯​ brain​ ​​​ ​​ ​ ​​ ​​​​ http://advances.sciencemag.org/ the proportionality of ancestral-to-descendant change in mean of stem toothed whales is 0.50:0.58. The ratio ( ¯​ body​ ) is 0.19, which brain size relative to mean body size. The scaling coefficient of the 0.39 below the upper bound of the brain-to-body relationship. ¯body​ ​ ​​​ ​​ ​ _ ​​ ​​​​ brain-to-body relationship of the ancestral grade was used as the The ratio ( ¯​ brain)​ is 5.39, which 3.65 above the upper bound of the expected proportionality of this ratio. The upper bound of the 95% body-to-brain relationship. Therefore, it is inferred that stem toothed confidence interval of the scaling coefficient of the ancestral grade whales indicated more change in mean body size than change in was used as the cutoff to infer that the change in mean size observed mean brain size than allometrically expected given their ancestral from ancestral-to-descendant grade is characterized by more change grade (specifically, more decrease in mean body size than decrease in mean brain size than body size. To account for the fact that gen­ in mean brain size). eralized least-squares procedures minimize residual error for the It should be noted that this procedure is valid only in the case of dependent variable, we inverse this procedure when evaluating a positive brain-to-body correlation. This assumption is not upheld changes in mean¯ body size. For body size, we thus considered the in delphinids and hominins, who demonstrate a decrease in body size body​ ​ on April 28, 2021 ​​​ ​​ ​ _ ​ ​​ proportion ( ¯brain​ ​) ​​ relative to the scaling coefficient of the ancestral and an increase in brain size (table S3). It is, however, evident that body-to-brain relationship (table S4). Although we consider the selection favors increased brain size relative to body size in these cases. use of the body-to-brain relationship for evaluating body size to be more rigorous than also using the brain-to-body relationship for Assessing differential changes in the variance of brain and/ these purposes, we emphasize that the results are largely unaffected or body size by this choice (i.e., the same results regarding disproportionate Patterns of changes in the variance of brain size and body size brain/body increase/decrease are attained when using the brain-­ among grades were evaluated by comparing the differences in variances to-body relationship to evaluate both brain size and body size). in brain size and body size between ancestral and descendant grades More change in mean brain size than mean body size is inferred [phylogenetic variance was calculated following standard phylogenetic ¯brain​ ​ ​​​ ​​ ​_ ​​ ​​​​ when ( ¯body​ ​) is higher than ¯the upper bound of the ancestral brain-­ generalized least-squares procedures (20)]. The change in ancestral-­ _​ body​ to-body expectation and ​(​​ ​​ ​ ¯​ brain​) ​​ is lower than the upper bound body- to-descendant body size variance is expected to be 1:1 for all grades, to-brain expectation. In Fig. 1, this scenario is indicated as two as this would maintain the proportionality of scaling differences from arrows for mean brain size and one arrow for mean body size. More ancestral-to-descendant grades. If this ratio is >1, then changes in change in mean body size than mean brain size is inferred when body size variance are greater than changes in brain size variance. If ¯brain​ ​ ​​​ _ ​​ ​​​​ ( ¯body​ ​) is lower than¯ the upper bound of the ancestral brain-to-body this ratio is <1, then changes in brain size variance are greater than _​ body​ expectation and (​​​ ¯​ brain ​) ​​ is higher than the upper bound body-to- changes in body size variance. Results are presented in table S5. brain expectation. In Fig. 1, this scenario is indicated as one arrow for mean brain size and two arrows for mean body size (which is the case SUPPLEMENTARY MATERIALS only for toothed whales). In pinnipeds, the observed proportion lies Supplementary material for this article is available at http://advances.sciencemag.org/cgi/ above the upper bound of both the ancestral brain-to-body and body- content/full/7/18/eabe2101/DC1 to-brain relationship. This is therefore indicated as two arrows for View/request a protocol for this paper from Bio-protocol. brain size and two for body size (both indicating an increase in size). For example, stem cercopithecoid primates (consisting of the REFERENCES AND NOTES fossil Victoriapithecus and extant colobines) derive directly from 1. J. F. Eisenberg, D. E. Wilson, Relative brain size and feeding strategies in the Chiroptera. the mammalian ancestral grade (Fig. 1). This gives this grade an Evolution 32, 740–751 (1978).

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 8 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE

2. S. Shultz, R. Dunbar, Both social and ecological factors predict ungulate brain size. 35. B. L. Finlay, R. B. Darlington, Linked regularities in the development and evolution Proc. R. Soc. Lond. B Biol. Sci. 273, 207–215 (2006). of mammalian brains. Science 268, 1578–1584 (1995). 3. D. Sol, R. P. Duncan, T. M. Blackburn, P. Cassey, L. Lefebvre, Big brains, enhanced 36. J. B. Smaers, A. Gómez-Robles, A. N. Parks, C. C. Sherwood, Exceptional evolutionary cognition, and response of birds to novel environments. Proc. Natl. Acad. Sci. 102, expansion of prefrontal cortex in great apes and humans. Curr. Biol. 27, 714–720 (2017). 5460–5465 (2005). 37. R. A. Barton, P. H. Harvey, Mosaic evolution of brain structure in mammals. Nature 405, 4. R. D. Martin, Relative brain size and basal metabolic rate in terrestrial vertebrates. Nature 1055–1058 (2000). 293, 57–60 (1981). 38. J. B. Smaers, A. H. Turner, A. Gómez-Robles, C. C. Sherwood, A cerebellar substrate 5. M. Kleiber, The Fire of Life: An Introduction to Animal Energetics (John Wiley&Sons, Inc., 1961). for cognition evolved multiple times independently in mammals. eLife 7, e35696 (2018). 6. E. Dubois, Sur le rapport du poids de l'encéphale avec la grandeur du corps chez les 39. R. J. Schusterman, C. R. Kastak, D. Kastak, The cognitive sea lion: Meaning and memory in mammifères. Bull. Mém. Soc. Anthropol. Paris 8, 337–376 (1897). the laboratory and in nature, in The Cognitive Animal: Empirical and Theoretical 7. H. J. Jerison, Evolution of the Brain and Intelligence (Academic Press, New York, 1973). Perspectives on Animal Cognition, M. Bekoff, C. Allen, G. M. Burghardt, Eds. (The MIT Press, 8. H. J. Jerison, Animal intelligence as encephalization. Philos. Trans. R. Soc. Lond B Biol. Sci. Cambridge, MA, 2002), pp. 217–228. 308, 21–35 (1985). 40. R. E. Passingham, J. B. Smaers, Is the prefrontal cortex especially enlarged in the human 9. C. C. Gilbert, W. L. Jungers, Comment on relative brain size in early primates and the use brain? Allometric relations and remapping factors. Brain Behav. Evol. 84, 156–166 (2014). of encephalization quotients in primate evolution. J. H um. Evol. 109, 79–87 (2017). 41. J. B. Smaers, D. R. Vanier, Brain size expansion in primates and humans is explained by 10. J. B. Smaers, C. S. Mongle, K. Safi, D. K. Dechmann, Allometry, evolution and development a selective modular expansion of the cortico-cerebellar system. Cortex 118, 292–305 of neocortex size in mammals. Prog. Brain Res. 250, 83–107 (2019). (2019). 11. J. R. Burger, M. A. George Jr., C. Leadbetter, F. Shaikh, The allometry of brain size 42. L. Marino, Big brains do matter in new environments. Proc. Natl. Acad. Sci. U.S.A. 102, in mammals. J. Mammal. 100, 276–283 (2019). 5306–5307 (2005). 12. D. T. Ksepka, A. M. Balanoff, N. A. Smith, G. S. Bever, B. A. S. Bhullar, E. Bourdon, E. L. Braun, 43. L. Marino, Cetacean brain evolution: Multiplication generates complexity. Int. J. Comp. J. G. Burleigh, J. A. Clarke, M. W. Colbert, J. R. Corfield, F. J. Degrange, V. L. de Pietri, Psychol. 17, 1–16 (2004).

C. M. Early, D. J. Field, P. M. Gignac, M. E. L. Gold, R. T. Kimball, S. Kawabe, L. Lefebvre, 44. A. Navarrete, C. P. van Schaik, K. Isler, Energetics and the evolution of human brain size. Downloaded from J. Marugán-Lobón, C. S. Mongle, A. Morhardt, M. A. Norell, R. C. Ridgely, R. S. Rothman, Nature 480, 91–93 (2011). R. P. Scofield, C. P. Tambussi, C. R. Torres, M. van Tuinen, S. A. Walsh, A. Watanabe, 45. X. Liu, M. Somel, L. Tang, Z. Yan, X. Jiang, S. Guo, Y. Yuan, L. He, A. Oleksiak, Y. Zhang, N. Li, L. M. Witmer, A. K. Wright, L. E. Zanno, E. D. Jarvis, J. B. Smaers, Tempo and pattern Y. Hu, W. Chen, Z. Qiu, S. Paabo, P. Khaitovich, Extension of cortical synaptic development of avian brain size evolution. Curr. Biol. 30, 2026–2036.e3 (2020). distinguishes humans from chimpanzees and macaques. Genome Res. 22, 611–622 (2012). 13. G. G. Simpson, Tempo and Mode in Evolution (Columbia Univ. Press, 1944). 46. M. Somel, H. Franz, Z. Yan, A. Lorenc, S. Guo, T. Giger, J. Kelso, B. Nickel, M. Dannemann, 14. C. Pélabon, C. Firmat, G. H. Bolstad, K. L. Voje, D. Houle, J. Cassara, A. L. Rouzic, S. Bahn, M. J. Webster, C. S. Weickert, M. Lachmann, S. Paabo, P. Khaitovich, Ann. N. Y. Acad. Sci. 1320 T. F. Hansen, Evolution of morphological allometry. , 58–75 Transcriptional neoteny in the human brain. Proc. Natl. Acad. Sci. U.S.A. 106, 5743–5748 http://advances.sciencemag.org/ (2014). (2009). 15. M. Tsuboi, W. van der Bijl, B. T. Kopperud, J. Erritzøe, K. L. Voje, A. Kotrschal, K. E. Yopak, 47. R. I. M. Dunbar, S. Shultz, Evolution in the social brain. Science 317, 1344–1347 (2007). S. P. Collin, A. N. Iwaniuk, N. Kolm, Breakdown of brain–body allometry 48. A. N. Iwaniuk, J. E. Nelson, Can endocranial volume be used as an estimate of brain size and the encephalization of birds and mammals. Nat. Ecol. Evol. 2, 1492–1500 (2018). in birds? Can. J. Zool. 80, 16–23 (2002). 16. A. Goswami, J. B. Smaers, C. Soligo, P. D. Polly, The macroevolutionary consequences 49. C. J. Logan, T. H. Clutton-Brock, Validating methods for estimating endocranial volume of phenotypic integration: From development to deep time. Philos. Trans. R. Soc. B Biol. in individual red deer (Cervus elaphus). Behav. Processes 92, 143–146 (2013). Sci. 369 , 20130254 (2014). 50. S. Faurby, J.-C. Svenning, A species-level phylogeny of all extant and late quaternary 17. J. A. Finarelli, J. J. Flynn, Brain-size evolution and sociality in Carnivora. Proc. Natl. Acad. extinct mammals using a novel heuristic-hierarchical Bayesian approach. Mol. Phylogenet. Sci. U.S.A. 106, 9345–9349 (2009). Evol. 84, 14–26 (2015). 18. V. Weisbecker, A. Goswami, Marsupials indeed confirm an ancestral mammalian pattern: 51. M. A. Butler, A. A. King, Phylogenetic comparative analysis: A modeling approach A reply to Isler. BioEssays 33, 358–361 (2011). for adaptive evolution. Am. Natural. 164, 683–695 (2004). 19. J. C. Uyeda, L. J. Harmon, A novel bayesian method for inferring and interpreting the dynamics 52. L. S. T. Ho, C. Ané, Intrinsic inference difficulties for trait evolution with Ornstein- of adaptive landscapes from phylogenetic comparative data. Syst. Biol. 63, 902–918 (2014). Uhlenbeck models. Methods Ecol. Evol. 5, 1133–1146 (2014). on April 28, 2021 20. J. B. Smaers, F. J. Rohlf, Testing species’ deviation from allometric predictions using 53. S. H. Montgomery, J. H. Geisler, M. R. McGowen, C. Fox, L. Marino, J. Gatesy, The the phylogenetic regression. Evolution 70, 1145–1149 (2016). evolutionary history of cetacean brain and body size. Evolution 67, 3339–3353 (2013). 21. J. C. Uyeda, M. W. Pennell, E. T. Miller, R. Maia, C. R. McClain, The evolution of energetic 54. L. Marino, D. W. McShea, M. D. Uhen, Origin and evolution of large brains in toothed scaling across the vertebrate tree of life. Am. Nat. 190, 185–199 (2017). whales. Anat. Rec. A Discov. Mol. Cell. Evol. Biol. 281, 1247–1255 (2004). 22. G. F. Striedter, Principles of Brain Evolution (Sinauer Associates, 2005). 55. J. B. Smaers, C. S. Mongle, On the accuracy and theoretical underpinnings of the multiple 23. J. G. Fleagle, R. A. Mittermeier, Locomotor behavior, body size, and comparative ecology variance Brownian motion approach for estimating variable rates and inferring ancestral of seven Surinam monkeys. Am. J. Phys. Anthropol. 52, 301–314 (1980). states. Biol. J. Linn. Soc. 121, 229–238 (2017). 24. W. Gearty, C. R. McClain, J. L. Payne, Energetic tradeoffs control the size distribution 56. J. B. Smaers, C. S. Mongle, A. Kandler, A multiple variance Brownian motion framework of aquatic mammals. Proc. Natl. Acad. Sci. U.S.A. 115, 4194–4199 (2018). for estimating variable rates and inferring ancestral states. Biol. J. Linn. Soc. 118, 78–94 25. R. Byrne, A. Whiten, Machiavellian Intelligence: Social Expertise and the Evolutoin of Intellect (2016). in Monkeys, Apes, and Humans (Oxford Univ. Press Inc., New York, 1988). 57. M. Khabbazian, R. Kriebel, K. Rohe, C. Ané, Fast and accurate detection of evolutionary 26. A. R. DeCasien, S. A. Williams, J. P. Higham, Primate brain size is predicted by diet but not shifts in Ornstein–Uhlenbeck models. Methods Ecol. Evol. 7, 811–824 (2016). sociality. Nat. Ecol. Evol. 1, 0112 (2017). 58. M. Archer, F. Jenkins Jr, S. Hand, P. Murray, H. Godthelp, Description of the skull and 27. J. R. Speakman, D. W. Thomas, Physiological ecology and energetics of bats, in Bat non-vestigial dentition of a Miocene Platypus (Obdurodon dicksoni, n. sp.) from Ecology, T. Kunz, M. Fenton, Eds. (University of Chicago Press, 2003), pp. 430–490. Riversleigh, Australia, and the problem of monotreme origins, in Platypus and Echidnas, 28. K. P. Dial, Evolution of avian locomotion: Correlates of flight style, locomotor modules, M. Augee, Ed. (Royal Zoological Society of New South Wales, Australia, 1993), pp. 15–27. nesting biology, body size, development, and the origin of flapping flight. The Auk 120, 59. T. E. Macrini, T. Rowe, M. Archer, Description of a cranial endocast from a fossil platypus, 941–952 (2003). Obdurodon dicksoni (Monotremata, Ornithorhynchidae), and the relevance of endocranial 29. M. Nakatsukasa, Y. Kunimatsu, Nacholapithecus and its importance for understanding characters to monotreme monophyly. J. Morphol. 267, 1000–1015 (2006). hominoid evolution. Evol. Anthropol. Issues News Rev. 18, 103–119 (2009). 60. K. W. S. Ashwell, Encephalization of Australian and new guinean marsupials. Brain Behav. 30. L. Marino, D. Sol, K. Toren, L. Lefebvre, Does diving limit brain size in cetaceans? Mar. Mamm. Sci. Evol. 71, 181–199 (2008). 22, 413–425 (2006). 61. A. M. Forasiepi, R. D. Macphee, S. H. del Pino, Caudal Cranium of Thylacosmilus atrox 31. T. Stankowich, A. N. Romero, The correlated evolution of antipredator defences and brain (Mammalia, Metatheria, Sparassodonta), a South American Predaceous Sabertooth. size in mammals. Proc. R. Soc. B Biol. Sci. 284, 20161857 (2017). Bull. Am. Museum Natural His. 2019, 1–66 (2019). 32. M. Genoud, Energetic strategies of shrews: Ecological constraints and evolutionary 62. J. Quiroga, M. Dozo, The brain of Thylacosmilus atrox. Extinct South American implications. Mammal Rev. 18, 173–193 (1988). saber-tooth carnivore marsupial. J. Hirnforsch. 29, 573–586 (1988). 33. S. D. Healy, C. Rowe, Costs and benefits of evolving a larger brain: Doubts over 63. P. S. Tambusso, R. A. Fariña, Digital cranial endocast of Pseudoplohophorus absolutus the evidence that large brains lead to better cognition. Anim. Behav. 86, e1–e3 (2013). (Xenarthra, Cingulata) and its systematic and evolutionary implications. J. Vertebr. 34. H. J. Jerison, Brain to body ratios and the evolution of intelligence. Science 121, 447–449 (1955). Paleontol. 35, e967853 (2015).

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 9 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE

64. Z. Kielan-Jaworowska, Evolution of the therian mammals in the Late Cretaceous of Asia. 91. R. D. Martin, A.-E. Martin, Primate Origins and Evolution: A Phylogenetic Reconstruction Part VI. Endocranial casts of eutherian mammals. Palaeontol. Polonica 46, 157–171 (1984). (Princeton Univ. Press Princeton, 1990). 65. J. Benoit, N. Crumpton, S. Mérigeaud, R. Tabuce, A memory already like an Elephant’s? 92. S. H. Montgomery, I. Capellini, R. A. Barton, N. I. Mundy, Reconstructing the ups The advanced brain morphology of the last common ancestor of afrotheria (Mammalia). and downs of primate brain evolution: Implications for adaptive hypotheses and Homo Brain Behav. Evol. 81, 154–169 (2013). floresiensis. BMC Biol. 8, 9 (2010). 66. K. Isler, C. P. van Schaik, Allomaternal care, life history and brain size evolution 93. W. C. Parr, H. J. Chatterjee, C. Soligo, Inter- and intra-specific scaling of articular surface in mammals. J. Hum. Evol. 63, 52–63 (2012). areas in the hominoid talus. J. Anat. 218, 386–401 (2011). 67. J. Benoit, L. J. Legendre, R. Tabuce, T. Obada, V. Mararescul, P. Manger, Brain 94. G. T. Schwartz, P. Mahoney, L. R. Godfrey, F. P. Cuozzo, W. L. Jungers, G. F. N. Randria, evolution in Proboscidea (Mammalia, Afrotheria) across the Cenozoic. Sci. Rep. 9, Dental development in Megaladapis edwardsi (Primates, Lemuriformes): Implications 9323 (2019). for understanding life history variation in subfossil lemurs. J. Hum. Evol. 49, 702–721 68. D. P. Domning, Encyclopedia of Marine Mammals, B. Würsig, J. Thewissen, K. M. Kovacs, (2005). Eds. (Elsevier, London, United Kingdom, 2018). 95. K. E. Sears, J. A. Finarelli, J. J. Flynn, A. R. Wyss, Estimating body mass in New World 69. P. Pirlot, T. Kamiya, Qualitative and quantitative brain morphology in the Sirenian “monkeys” (Platyrrhini, Primates), with a consideration of the Miocene platyrrhine, Dugong dugong Erxl. J. Zoolog. Syst. Evol. Res. 23, 147–155 (1985). Chilecebus carrascoensis. Am. Museum Novitates 3617, 1–29 (2008). 70. R. Savage, D. P. Domning, J. Thewissen, Fossil Sirenia of the West Atlantic and Caribbean 96. M. T. Silcox, C. K. Dalmyn, J. I. Bloch, Virtual endocast of Ignacius graybullianus region. V. The most primitive known sirenian,Prorastomus sirenoides Owen, 1855. (Paromomyidae, Primates) and brain evolution in early primates. Proc. Natl. Acad. Sci. J. Vert. Paleontol. 14, 427–449 (1994). U.S.A. 106, 10987–10992 (2009). 71. J. Shoshani, W. J. Kupsky, G. H. Marchant, Elephant brain: Part I: Gross morphology, 97. E. L. Simons, The cranium of Parapithecus grangeri, an Egyptian Oligocene anthropoidean functions, comparative anatomy, and evolution. Brain Res. Bull. 70, 124–157 (2006). primate. Proc. Natl. Acad. Sci. U.S.A. 98, 7892–7897 (2001). 72. E. Palkopoulou, M. Lipson, S. Mallick, S. Nielsen, N. Rohland, S. Baleka, E. Karpinski, 98. S. G. Strait, Dietary reconstruction of small-bodied omomyoid primates. J. Vert. Paleontol. A. M. Ivancevic, T. H. To, R. D. Kortschak, J. M. Raison, Z. Qu, T. J. Chin, K. W. Alt, 21, 322–334 (2001).

S. Claesson, L. Dalén, R. D. E. MacPhee, H. Meller, A. L. Roca, O. A. Ryder, D. Heiman, 99. J. Orihuela, Endocranial morphology of the extinct Antillean shrew Nesophontes Downloaded from S. Young, M. Breen, C. Williams, B. L. Aken, M. Ruffier, E. Karlsson, J. Johnson, F. di Palma, (Lipotyphla: Nesophontidae) from natural and digital endocasts of Cuban taxa. J. Alfoldi, D. L. Adelson, T. Mailund, K. Munch, K. Lindblad-Toh, M. Hofreiter, H. Poinar, Palaeontol. Electron. 17, 153 (2014). D. Reich, A comprehensive genomic history of extinct and living elephants. Proc. Natl. 100. G. Baron, H. Stephan, H. D. Frahm, Comparative Neurobiology in Chiroptera: Acad. Sci. U.S.A 115, E2566–E2574 (2018). Macromorphology, Brain Structures, Tables and Atlases (Birkhäuser, Basel, 1996), vol. 1. 73. F. Delsuc, G. C. Gibb, M. Kuch, G. Billet, L. Hautier, J. Southon, J. M. Rouillard, 101. U. Norberg, M. Fenton, P. Racey, J. Rayner (Cambridge Univ. Press Cambridge, 1987). J. C. Fernicola, S. F. Vizcaíno, R. D. E. MacPhee, H. N. Poinar, The phylogenetic affinities 102. J. A. Simmons, Comparative Neurobiology in Chiroptera. Science 273, 609–610 (1996). Am. Nat. 112 of the extinct glyptodonts. Curr. Biol. 26, R155–R156 (2016). 103. L. Radinsky, Evolution of brain size in carnivores and ungulates. , 815–831 http://advances.sciencemag.org/ 74. S. Harel, K. Watanabe, I. Linke, R. J. Schain, Growth and development of the rabbit brain. (1978). Neonatology 21, 381–399 (1972). 104. M. Spaulding, M. A. O'Leary, J. Gatesy, Relationships of Cetacea (Artiodactyla) among 75. S. Herculano-Houzel, P. Ribeiro, L. Campos, A. Valotta da Silva, L. B. Torres, K. C. Catania, mammals: Increased taxon sampling alters interpretations of key fossils and character J. H. Kaas, Updated neuronal scaling rules for the brains of Glires (rodents/lagomorphs). evolution. PloS ONE 4, e7062 (2009). Brain Behav. Evol. 78, 302–314 (2011). 105. A. Zanazzi, M. J. Kohn, D. O. Terry Jr., Biostratigraphy and paleoclimatology of the Eocene- 76. P. Pirlot, T. Kamiya, Relative size of brain and brain components in three gliding Oligocene boundary section at Toadstool Park, northwestern Nebraska, USA. Geol. Soc. placentals (Dermoptera: Rodentia). Can. J. Zool. 60, 565–572 (1982). Am. 452, 197–214 (2009). 77. A. Beaudet, J. Dumoncel, F. de Beer, B. Duployer, S. Durrleman, E. Gilissen, J. Hoffman, 106. J. H. Geisler, M. R. McGowen, G. Yang, J. Gatesy, A supermatrix analysis of genomic, BMC Evol. Biol. 11 C. Tenailleau, J. F. Thackeray, J. Braga, Morphoarchitectural variation in South African morphological, and paleontological data from crown Cetacea. , 112 fossil cercopithecoid endocasts. J. Hum. Evol. 101, 65–78 (2016). (2011). 78. J. I. Bloch, M. T. Silcox, D. M. Boyer, E. J. Sargis, New Paleocene skeletons 107. G. T. Lloyd, G. J. Slater, A total-group phylogenetic metatree of Cetacea and the System. Biol. 2021 and the relationship of plesiadapiforms to crown-clade primates. Proc. Natl. Acad. Sci. importance of fossil data in diversification analyses. , syab002 (2021). U.S.A. 104, 1159–1164 (2007). 108. P. R. Manger, An examination of cetacean brain structure with a novel hypothesis on April 28, 2021 Biol. Rev. 81 79. D. A. Burney, L. P. Burney, L. R. Godfrey, W. L. Jungers, S. M. Goodman, H. T. Wright, correlating thermogenesis to the evolution of a big brain. , 293–338 (2006). A. J. T. Jull, A chronology for late prehistoric Madagascar. J. Hum. Evol. 47, 25–63 109. L. Marino, R. C. Connor, R. E. Fordyce, L. M. Herman, P. R. Hof, L. Lefebvre, D. Lusseau, (2004). B. McCowan, E. A. Nimchinsky, A. A. Pack, L. Rendell, J. S. Reidenberg, D. Reiss, M. D. Uhen, 80. E. Delson, C. J. Terranova, W. L. Jungers, E. J. Sargis, N. G. Jablonski, Body Mass in E. van der Gucht, H. Whitehead, Cetaceans have complex brains for complex cognition. PLOS Biol. 5 Cercopithecidae (Primates, Mammalia): Estimation and Scaling in Extinct and Extant Taxa, , e139 (2007). (Anthropological Papers, American Museum of Natural History, 2000), vol. 83. 110. S. H. Ridgway, A. C. Hanson, Sperm whales and killer whales with the largest brains of all toothed whales show extreme differences in cerebellum. Brain Behav. Evol. 83, 266–274 81. J. Gómez, M. Verdú, Mutualism with plants drives primate diversification. Syst. Biol. 61, (2014). 567–577 (2012). 111. M. D. Uhen, Form, Function, and Anatomy of Dorudon Atrox (Mammalia, Cetacea): An 82. L. A. Gonzales, B. R. Benefit, M. L. McCrossin, F. Spoor, Cerebral complexity preceded Archaeocete from the Middle to Late Eocene of Egypt (University of Michingan, 2004). enlarged brain size and reduced olfactory bulbs in Old World monkeys. Nat. Commun. 6, 112. L. Radinsky, Oldest horse brains: More advanced than previously realized. Science 194, (2015). 626–627 (1976). 83. M. Grabowski, K. G. Hatala, W. L. Jungers, B. G. Richmond, Body mass estimates 113. M. Dubied, F. Solé, B. Mennecart, The cranium of Proviverra typica (Mammalia, of hominin fossils and the evolution of human body size. J. Hum. Evol. 85, 75–93 (2015). Hyaenodonta) and its impact on hyaenodont phylogeny and endocranial evolution. 84. M. Grabowski, W. L. Jungers, Evidence of a chimpanzee-sized ancestor of humans but Palaeontology 62, 983–1001 (2019). a gibbon-sized ancestor of apes. Nat. Commun. 8, 880 (2017). 114. R. Hunt, in The George Wright Forum (JSTOR, 1984), vol. 4, pp. 29–39. 85. L. B. Halenar, S. B. Cooke, A. L. Rosenberger, R. Rímoli, New cranium of the endemic 115. S. Tomiya, Z. J. Tseng, Whence the beardogs? Reappraisal of the Middle to Late Eocene Caribbean platyrrhine, Antillothrix bernensis, from La Altagracia Province, Dominican ‘Miacis’ from Texas, USA, and the origin of Amphicyonidae (Mammalia, Carnivora). Republic. J. Human Evol. 106, 133–153 (2017). R. Soc. Open Sci. 3, 160518 (2016). 86. K. Isler, E. Christopher Kirk, J. M. A. Miller, G. A. Albrecht, B. R. Gelvin, R. D. Martin, Endocranial volumes of primate species: Scaling analyses using a comprehensive and reliable data set. J. Hum. Evol. 55, 967–978 (2008). 87. J. Kappelman, The evolution of body mass and relative brain size in fossil hominids. Acknowledgments: We thank E. R. Seiffert for the useful comments and discussion, and J. Hum. Evol. 30, 243–276 (1996). K. W. S. Ashwell for sharing marsupial data. We further thank J. Lázaro for making Fig. 1. 88. E. C. Kirk, E. L. Simons, Diets of fossil primates from the Fayum Depression of Egypt: Funding: J.B.S. was funded by the National Science Foundation (grant 80692). A.M.B. was A quantitative analysis of molar shearing. J. Hum. Evol. 40, 203–229 (2001). funded by the National Science Foundation (grant DEB 1801224). A.G. was funded by the 89. L. Kordos, New results of Hominoid research in the Carpathian Basin. Acta Biol. European Research Council (H2020 ERC-Stg-637171). C.S.M. was supported by the Szegediensis 44, 71–74 (2000). Gerstner Fellowship and the Gerstner Family Foundation, the Kalbfleisch Fellowship, and 90. L. Marino, A comparison of encephalization between odontocete cetaceans the Richard Gilder Graduate School of the American Museum of Natural History. V.W. was and anthropoid primates. Brain Behav. Evol. 51, 230–238 (1998). funded by the Australian Research Council Discovery Grant (DP170103227). D.d.V. was

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 10 of 11 SCIENCE ADVANCES | RESEARCH ARTICLE supported by funds from the Natural Environment Research Council (NERC NE/ Submitted 6 August 2020 T000341/1). Author contributions: J.B.S., D.K.N.D., K.S., R.S.R., and D.R.H. gleaned data Accepted 10 March 2021 from the literature. D.R.H., B.B., J.G.F., C.C.G., A.G., W.L.J., P.R.M., and C.S.M. assisted with Published 28 April 2021 phylogenetic placement of extinct species. J.B.S., R.S.R., and D.R.H. performed the 10.1126/sciadv.abe2101 statistical analyses. J.B.S. wrote the first draft. All authors read and edited the paper. Competing interests: The authors declare that they have no competing interests. Data Citation: J. B. Smaers, R. S. Rothman, D. R. Hudson, A. M. Balanoff, B. Beatty, D. K. N. Dechmann, and materials availability: All data needed to evaluate the conclusions in the paper are D. de Vries, J. C. Dunn, J. G. Fleagle, C. C. Gilbert, A. Goswami, A. N. Iwaniuk, W. L. Jungers, present in the paper and/or the Supplementary Materials. Additional data related to this M. Kerney, D. T. Ksepka, P. R. Manger, C. S. Mongle, F. J. Rohlf, N. A. Smith, C. Soligo, V. Weisbecker, paper may be requested from the authors. K. Safi, The evolution of mammalian brain size. Sci. Adv. 7, eabe2101 (2021). Downloaded from http://advances.sciencemag.org/ on April 28, 2021

Smaers et al., Sci. Adv. 2021; 7 : eabe2101 28 April 2021 11 of 11 The evolution of mammalian brain size J. B. Smaers, R. S. Rothman, D. R. Hudson, A. M. Balanoff, B. Beatty, D. K. N. Dechmann, D. de Vries, J. C. Dunn, J. G. Fleagle, C. C. Gilbert, A. Goswami, A. N. Iwaniuk, W. L. Jungers, M. Kerney, D. T. Ksepka, P. R. Manger, C. S. Mongle, F. J. Rohlf, N. A. Smith, C. Soligo, V. Weisbecker and K. Safi

Sci Adv 7 (18), eabe2101. DOI: 10.1126/sciadv.abe2101 Downloaded from

ARTICLE TOOLS http://advances.sciencemag.org/content/7/18/eabe2101

SUPPLEMENTARY http://advances.sciencemag.org/content/suppl/2021/04/26/7.18.eabe2101.DC1

MATERIALS http://advances.sciencemag.org/

REFERENCES This article cites 99 articles, 14 of which you can access for free http://advances.sciencemag.org/content/7/18/eabe2101#BIBL

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions on April 28, 2021

Use of this article is subject to the Terms of Service

Science Advances (ISSN 2375-2548) is published by the American Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005. The title Science Advances is a registered trademark of AAAS. Copyright © 2021 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of Science. No claim to original U.S. Government Works. Distributed under a Creative Commons Attribution NonCommercial License 4.0 (CC BY-NC).